ABSTRACT Substance use disorders (SUD) result in increased risk of morbidity and mortality and impose substantial personal and financial burdens on affected individuals and society as a whole. There is considerable evidence that genetics increases the risk for SUDs, but identifying specific gene variants has been hampered by disease heterogeneity, genetic complexity and the inability to control or account for environmental exposures in human populations. Some of these issues can be overcome with addiction genetic studies in rodent models. However, standard methods of mapping in rodent populations often result in identification of large genomic regions that require considerable effort to narrow and thus delay or prevent identification of the causal gene variant. Recent technology and bioinformatic resources have made it possible to conduct genetic mapping studies using crosses between inbred mouse substrains (called Reduced Complexity Crosses or RCCs). Inbred mouse substrains diverge when inbred strain breeding colonies are transferred between collaborators or commercial entities. Polymorphisms segregate between these subpopulations at regions of the genome for which the parental strain has not yet become fixed, or from genetic drift, resulting in sets of strains that are genetically very closely related. Therefore, while standard inbred strains differ from each other at hundreds of thousands of loci, inbred substrains differ at thousands of loci ? only a fraction of which will be functional. These characteristics expedite identification of causal genes in substrains for which phenotypic differences exist. This approach has been used to identify a cocaine-sensitivity gene but has been limited to C57BL/6 substrains. Dozens of inbred substrains exist and opportunities for identifying genes that influence addiction-related behaviors are being missed. We have identified a striking phenotypic difference in cocaine locomotor sensitivity between the C3H substrains, C3H/HeJ and C3H/HeNTac. In this R21 application, we propose to use a RCC to identify the causal allele(s) for the phenotypic differences in these substrains. We will also examine substrain-specific pharmacokinetic and dose response phenotypes and examine the rewarding and reinforcing effects of cocaine using the drug self- administration paradigm. These studies will establish a drug response profile for each C3H substrain and allow us to begin exploring mechanisms by which the causal allele alters behavior.